Lojban has no entry level material that is actually engaging. We need a large number of articles about interesting things, but written at the language level of a four year old. With nothing interesting to read, and no other reason to learn other than "it's interesting" it's having a hard time growing.
It needs entry level interesting material.
In fact, the strongest headwind Esperanto constantly faces is Spanish. There's already a romance language with absolutely systematic spelling, easy pronunciation, simple grammar, and few irregularities. It's already the second most widely spoken language on the planet, it has mountains of interesting and elementary material, and there are lots of interesting people whom you can meet only by speaking it.
It's hard for any rationalized romance language to gain traction against that.
Sure, that would be quite recursive. But you wouldn't read lojban if you weren't interested in it anyway, so it's bound to be interesting.
Lots of people hear of Esperanto and decide to pick it up. Almost no one hears of Lojban and picks up. Lots of people are fluent in Esperanto. Perhaps no one is fluent in Lojban. Granted Esperanto had a huge head-start, but I don't think you can attribute its relative success entirely to that factor.
Lojban is still missing something, despite being perfectly logical. English, on the other hand, is flourishing despite being extremely illogical. English is one of the most successful and least logical languages on Earth. Lojban is one of the least successful and most logical languages on Earth. "Correlation does not imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing 'look over there'." - Randall Munroe
All I was trying to say is that being logical is not a panacea.
There was once a lack of other people who spoke Klingon, and yet it flourished.
There was (at around the turn of the 20th century) a lack of people who spoke Hebrew, and yet it too flourished and now there's a whole nation of people who speak it.
Lack of people who speak the language is always a problem for any artificial language... at first. Some manage to overcome this problem. Others don't. The interesting question is why this is the case, and what can be done about it.